Estuaries and Coasts (2011) 34:386–390 DOI 10.1007/s12237-010-9330-1

NOTE

A Test of Reduction Devices on Commercial Crab Pots in a Tidal Marsh Creek in Virginia

A. Scott Morris & Stephanie M. Wilson & Erik F. Dever & Randolph M. Chambers

Received: 15 October 2009 /Revised: 5 July 2010 /Accepted: 7 July 2010 /Published online: 21 July 2010 # Coastal and Estuarine Research Federation 2010

Abstract The effectiveness of bycatch reduction devices declines in population sizes from various threats throughout (BRDs) on commercial pots designed to capture blue crabs their range have prompted many states to list terrapins as wastestedintheYorkRiveron endangered, threatened, or a species of special concern Felgates Creek (37.2667 N, −76.5850 W) over the period (Butler et al. 2006). Because the economies of many of 4 June through 31 July 2009. For each of 10 pairs of pots, these same states are bolstered by the harvest of the blue one had BRDs affixed to all four entrance gapes and the crab Callinectes sapidus, extensive use of commercial pots other had none. Pots were baited approximately once each to catch crabs has led to bycatch mortality of terrapins week but were sampled for blue crabs and bycatch 6 of (Wood 1997; Roosenburg and Green 2000; Hoyle and 7 days each week for the duration of the study. More than Gibbons 2000; Butler and Heinrich 2007; Dorcas et al. one fourth of 1,643 total crabs were caught on the first day 2007; Grosse et al. 2009; Rook et al. 2010). New York, after baiting, and for these 7 days, no statistical difference New Jersey, Maryland, and Delaware now require a version was detected between either the number or size of legal-size of a terrapin excluder device on certain commercial pots to crabs caught in BRD versus non-BRD pots. Of 51 reduce that source of mortality, but enforcement is difficult Malaclemys terrapin and44fishcaughtasbycatch (Roosenburg 2004). throughout the study, all but three fish were captured in Terrapins typically spend most of their lives in shallow non-BRD pots. BRDs exclude bycatch and may reduce water adjacent to tidal wetlands, so during the summer, incidental mortality of crabs in pots that are not tended only a small portion of the crab spatially intersects regularly. with turtle habitat (but see Roosenburg 2004). Within that shallow-water habitat, however, we recognize at least four Keywords Diamondback terrapin . Excluder . possible sources of commercial pots. First, some commer- Recreational crabbing cial hard-shell crabbers may run lines of pots along the mainstem of small tidal creeks, with daily baiting and harvesting. Second, commercial crabbers during a short Introduction period may run “peeler pots” baited with male crabs to attract female crabs ready to shed their shells prior to The diamondback terrapin Malaclemys terrapin is the only mating. Crabbers focus their effort in the relatively fully estuarine turtle in the Western Hemisphere, occurring protected habitat of small tidal creeks where peelers in coastal habitats from Massachusetts to the Gulf Coast of concentrate. Third, many recreational crabbers place one Texas (Ernst and Lovich 2009). Low numbers or measured or two commercial-type pots in shallow tidal waters near their waterfront property homes. Baiting and checking of : : : recreational pots may be infrequent. Finally, when hard- A. S. Morris S. M. Wilson E. F. Dever R. M. Chambers (*) shell crab pots placed in deeper water break free from their Biology Department and Keck Environmental Field Lab, moorings or are abandoned, they may be carried by storms College of William and Mary, Williamsburg, VA 23187, USA or tidal currents into shallow-water habitat. Though they are e-mail: [email protected] never baited, these ghost or derelict pots are potential, Estuaries and Coasts (2011) 34:386–390 387 chronic causes of bycatch mortality (Roosenburg 1991; gizzard shad (Dorosoma cepedianum) on average once each Guillory and Prejean 1998; Havens et al. 2008). week for 7 weeks during summer 2009. We chose this In most states where required, regulations to install bycatch experimental design because many recreational crabbers bait reduction devices (BRDs) on crab pots have targeted and/or check their pots infrequently; further, abandoned pots recreational crabbers, but clearly crab pots from other users are not tended at all, so the catch obtained after many days are either fished or end up in shallow-water habitat where without fresh baiting also is a potential proxy for pots that terrapins are most abundant. For the present study, we have been lost. The effects of self-baiting on crab capture by designed an experiment to test BRD function in a brackish abandoned pots, however, were not tested in this study. marsh complex. Our objective was to complete the first test of When sampling, every organism caught in a pot was BRDs in Virginia shallow subtidal waters where recreational, identified and recorded. Crabs were sexed, and point-to-point peeler, and derelict crab pots are located. By baiting pots at carapace width was measured using a sliding caliper. Legal- infrequent intervals, we were also able to assess the potential size crabs were those with carapace widths exceeding influence of some recreational pots (that are rebaited 12.5 cm. Each captured turtle was sexed and measured for infrequently) and derelict pots (that are never rebaited) on carapace width and shell depth. Finally, every newly captured crab catch and terrapin bycatch. turtle was marked by filing a unique series of notches in the marginal scutes (Cagle 1939); recaptures were noted. The number of fish in each pot was recorded and all live bycatch Materials and Methods was set free. Any dead fish remaining in pots were removed, and no bait remained in pots after the second day. Study Area Terrapin Population Estimate The experiment was conducted over the period beginning 4 June 2009 and ending 31 July 2009, with all sampling In addition to pots placed in the subtidal creeks, we placed ten completed in the upper reaches of Felgates Creek, a commercial pots fitted with 60-cm chimneys in small mesohaline tidal marsh creek connected to the York River, intertidal channels of the vegetated marsh. These pots on the Yorktown Naval Weapons Station (37.2667 remained unbaited throughout the same study period as the N, −76.5850 W). The surrounding wetland complex is BRD experiment and were sampled approximately 6 days dominated by the marsh grass Spartina alterniflora. each week of the study. All newly captured terrapins were Commercial and recreational crabbings have not been sexed and measured for carapace width and shell depth, allowed in Felgates Creek for over 40 years (Rook et al. marked with a unique code using the notching technique, and 2010). then released at the point of capture. Recaptures were noted, and population size was estimated using the Schnabel mark– Bycatch Experiment recapture method (Krebs 1989).

We placed ten pairs of commercial crab pots (galvanized steel; Data Analysis commercial and recreational pots require cull rings but these were not used in this study) in the subtidal portions of Felgates All capture data were tested for normality using Shapiro– Creek, to an approximate depth of 60 cm at mean low water so Wilk tests and found to be nonnormally distributed. that the pots were submerged at all times. This shallow Because no transformation led to normalization of the data, subtidal zone is where many recreational crab pots are placed nonparametric tests comparing crab catch and crab sizes in and where some derelict and abandoned pots get carried by BRD versus non-BRD pots were completed using Mann– wave and current action. To ensure that terrapins did not Whitney U tests. Bycatch differences as a function of bait drown when captured, all pots were fitted with sealed poultry day were tested using Kruskal–Wallis analysis of variance. wire “chimneys” that extended 120 cm above the top of the For all tests, the significance criterion was p<0.05. pot. During high tide, captured terrapins swam up the chimney to the water surface to breathe. For each pair of commercial pots, one pot was fitted with Results 12×4.5 cm BRDs (TopME© Products) attached near the interior end of the funnel of all four gapes; the other pot was Bycatch Experiment not fitted with BRDs. Our sampling strategy was designed to test the effect of BRDs and the impact of infrequent pot Throughout the 46 pot sampling days, we caught 1,634 baiting on crab catch and bycatch. Although sampling of pots total crabs, of which 679 were of legal size; male crabs was completed 6 days/week, we baited the pots with fresh comprised 93% of the total catch and 96% of the legal-size 388 Estuaries and Coasts (2011) 34:386–390 catch (Table 1). We caught a mean of 1.78±1.79 SD crabs Likewise, more legal-size crabs were caught on average in per pot per day, with a minimum of zero crabs and a pots without terrapin bycatch (1.49±1.46 versus 0.86±0.90 maximum of 11 crabs. Catch per unit effort (CPUE) for crabs), but the difference was not significant (Mann–Whitney total and legal-size crabs was highest on the first sampling U=170, p=0.306). date following pot baiting (Fig. 1). Over 25% of total crabs were caught on the first day after baiting (Table 1). For Terrapin Population Estimate these seven sampling days immediately after pot baiting, neither the difference in average ± SD CPUE (2.86±2.02 From the unbaited commercial pots placed and sampled in versus 3.33±2.26 crabs) nor legal-size CPUE (1.19±1.35 the intertidal channels of small marsh creeks, we caught versus 1.43±1.40) was significant between BRD and non- 143 terrapins, of which 98 were original captures and 45 BRD pots, respectively (Mann–Whitney U=2,213.5, p= were recaptures. The ratio of male to female terrapins was 0.318 and U=2,186.5, p=0.254). 2:1. Based on these data, we used the Schnabel mark– Across all other days since baiting (N=39), both the recapture method to estimate a total population size of 133 difference in total CPUE (1.07±1.16 versus 2.01±1.80) and terrapins with a 95% confidence interval between 102 and legal-size CPUE (0.44±0.70 versus 0.83±1.08) were 186 terrapins. Finally, we plotted terrapin shell depth and significant between BRD and non-BRD pots, respectively carapace width from all captures to show that based on size, (Mann–Whitney U=76,434.5, p<0.001 and U=86,733, 92% of all female and 70% of all male terrapins caught in p<0.001). The difference in mean ± SD size of legal-size non-BRD pots would have been excluded from pots fitted crabs in BRD pots (13.30±0.46 cm) and non-BRD pots with 12×4.5 cm BRDs (Fig. 2). That we caught no (13.42±0.68 cm), however, was not significant (U=51,155, terrapins in BRD-fitted pots despite some being small p=0.184). enough suggests possible avoidance behavior of terrapins to Although crab catch varied by the number of days since BRDs. pot baiting, the bycatch of terrapins was similar among sampling days irrespective of baiting (Kruskal–Wallis test, Χ2=11.71, df=12, p=0.514). All of the 51 terrapins caught Discussion as bycatch were in non-BRD pots, with 27 pots containing one terrapin, nine pots containing two terrapins, and two pots The use of BRDs on commercial crab pots placed in containing three terrapins. Further, we caught a total of 44 subtidal sections of marsh creeks had no statistically (Leiostomus xanthurus), Atlantic croaker (Micropogonias significant influence on crab catch the day after baiting. undulates), and summer flounder (Paralichthys dentatus), 41 This result is similar to the only other BRD study in of which were in non-BRD pots, and all but one of which Virginia waters (Rook et al. 2010), although for that study were caught beyond the first day after baiting. Similar to the pots were placed in more shallow, intertidal locations. terrapin bycatch, fish bycatch did not vary significantly With other BRD studies completed in New Jersey (Wood among bait days (Kruskal–Wallis test, Χ2=19.46, df=12, p= 1997), Maryland (Roosenburg and Green 2000), Florida 0.078). No other species comprised the bycatch. (Butler and Heinrich 2007), and Louisiana (Guillory and We compared the impact of terrapin bycatch on the Prejean 1998), the existing consensus is that crab catch is associated crab catch in non-BRD pots on the first day after not dramatically compromised by BRDs. baiting. For pots without terrapin bycatch (N=63), the total Beyond the first day after baiting, however, crab catch crab CPUE was 3.52±2.49 crabs; for pots in which one or dropped, demonstrating that crabs enter pots in search of more terrapins were caught (N=7), the total crab CPUE was food and are responding to the fresh bait. More importantly, only 1.57±1.27 crabs (Mann–Whitney U=112, p<0.05). when pots are not freshly baited, crabs do not enter BRD-

Table 1 Summary catch data for crab pots fitted with BRDs First day after baiting All other bait days Totals (ten pots) and those without BRDs (ten pots), splitting catch Catch BRD Non-BRD BRD Non-BRD from first day after baiting (N= 70 pot days) versus catch from Total crabs 200 233 419 782 1,634 all other days (N=380 pot days) Legal-size crabs 83 100 172 324 679 Males 189 210 396 720 1,515 Females 11 23 23 62 119 Terrapins 0 9 0 42 51 Fish 1 0 4 39 44 Estuaries and Coasts (2011) 34:386–390 389

a 10 Non BRD 1.6 Non-BRD BRD 8 Zone 6 1.2

4 Females Males 0.8 Shell Depth cm 2 BRD Zone

0 0.4

Legal crabs per pot 024681012141618 Carapace Width cm 0.0 Fig. 2 Plot of carapace width and shell depth for male and female b terrapins caught in non-BRD pots. No terrapins were captured in Non BRD BRD-fitted pots. Dimensions of BRDs used are also plotted to show BRD BRD zone, outside of which any terrapins captured in a non-BRD pot 3 would have been excluded based on size

been noted elsewhere (Roosenburg et al. 1997; Dorcas et al. 2 2007; Grosse et al. 2009). Interestingly, we found that terrapin and fish bycatch did not vary among bait days, i.e., terrapins and fish entered pots irrespective of whether bait 1 was fresh. On days that bait was fresh, however, the total Total crabs per pot crab catch was significantly lower in pots containing a terrapin. These results highlight two important concerns: (1) 0 1357911 unbaited pots pose an ongoing mortality threat to terrapins Days since baiting and fish and are at least as attractive as freshly baited pots and (2) the presence of live terrapins in pots reduces crab Fig. 1 a Mean number of legal-size crabs (+SE) as a function of days catch. Both of these concerns are addressed by BRDs that – since baiting pots with and without BRDs in June July 2009 in exclude terrapins, thereby reducing turtle mortality and Felgates Creek, Virginia; b mean total number of crabs (+SE) in pots with and without BRDs over the same period improving overall crab catch. The general impact of commercial crab pots on terrapin bycatch is fairly well-established (i.e., crab pots also fitted pots as frequently as non-BRD pots. Irregular tending capture turtles), but the relative contribution of different of pots might happen, for example, when some recreational crabbing groups (e.g., recreational versus commercial pots are visited only on weekends; derelict and abandoned crabber) to terrapin population dynamics is not as clear. pots are not tended at all and are a potential long-term sink We do not know to what extent the current distribution of for legal-size crabs (Havens et al. 2008). Although all diamondback terrapin populations has been influenced untended pots may self-bait over time, unbaited pots fitted by crabbing pressures, nor do we know about population with BRDs did not attract as many crabs or bycatch in the recovery when those crabbing pressures are reduced present study, suggesting that BRDs may reduce incidental through use of BRDs. Recent studies by Dorcas et al. capture and mortality of crabs when pots are not tended (2007) and Grosse et al. (2009) have documented long-term regularly. changes in population demographics and short-term deci- BRDs were highly effective at excluding both turtles and mation of population sizes, respectively, directly related to fish as bycatch. For male and female diamondback terrapins terrapin mortality in commercial crab pots. The benefits of in Felgates Creek, the total number caught (N=51) in the BRDs should be realized anywhere commercial-type crab ten non-BRD pots represents a potential reduction in pots are set or eventually settle into diamondback terrapin population size from 27% to 50%. Given that pots were habitat. sampled only 46 days, the terrapin population in this creek Because pots that are placed or are transported to tidal would have experienced significant mortality of adult males marsh creeks and other shallow subtidal areas (e.g., seagrass and juvenile and young adult females over a single full beds) may be owned by recreational crabbers, commercial season of commercial crabbing (8 months). Because the sex hard-shell crabbers, or peeler crabbers, regulations should not ratio of our bycatch of terrapins was male-biased (2:1), we necessarily target a specific crabbing group. Rather, we suggest would anticipate a shift in population demographics, as has that regulations should target the terrapin habitat, i.e., BRDs 390 Estuaries and Coasts (2011) 34:386–390 should be mandated for all crabbers in areas of potentially high Dorcas, M.E., J.D. Wilson, and J.W. Gibbons. 2007. Crab trapping terrapin density. Similar to current regulations in New Jersey, causes population decline and demographic changes in diamondback terrapins over two decades. Biological Conser- metrics such as size of tidal creek or adjacency to wetlands can vation 137: 334–340. be used to delineate potential terrapin habitat in other states. Ernst, C., and J. Lovich. 2009. Turtles of the United States and New Jersey requires that both commercial and recreational Canada. Baltimore, MD: Johns Hopkins University Press. pots within 150 ft (45.7 m) of the shoreline be fitted with BRDs Feinberg, J.A., and R.L. Burke. 2003. Nesting ecology and predation of diamondback terrapins, Malaclemys terrapin,at (Roosenburg 2004). Based on these metrics, maps of critical Gateway National Recreation Area, New York. Journal of habitat areas where BRDs are required could be posted, for Herpetology 37: 517–526. example, on a relevant agency website. Unfortunately, the Grosse, A.M., J.D. van Dijk, K.L. Holcomb, and J.C. Maerz. 2009. current distribution and sizes of terrapin populations relative Diamondback terrapin mortality in crab pots in a Georgia tidal marsh. Chelonian Conservation and Biology 8: 98–100. to potential habitat is an unknown and probably complex Guillory, V., and P. Prejean. 1998. Effect of a terrapin excluder device mosaic owing to slow, long-term recovery from species on blue crab, Callinectes sapidus, trap catches. Marine decimation occurring a century ago (Schaffer et al. 2008), 60: 38–40. more recent threats to successful nesting and juvenile Havens, K.J., D. Bilkovic, D. Stanhope, K. Angstadt, and C. Hershner. 2008. The effects of derelict blue crab traps on survivorship (Burger 1977; Feinberg and Burke 2003), and marine organisms in the lower York River, Virginia. North periodic pulses of adult mortality, primarily from drowning in American Journal of Fisheries Management 28: 1194–1200. commercial crab pots (Seigel and Gibbons 1995). Hoyle, M.E., and J.W. Gibbons. 2000. Use of a marked population of Conservation of terrapins by either protecting existing diamondback terrapins (Malaclemys terrapin) to determine impacts of recreational crab pots. Chelonian Conservation and populations or facilitating the recovery of others would be Biology 3: 735–737. best served by minimizing the potential impact from the Krebs, C.J. 1989. Ecological methodology, 2nd ed. New York: blue crab fishery in those areas where terrapins might live. Harper Collins. Although the BRD requirement is difficult to enforce, the Rook, M.A., R.L. Lipcius, B.M. Bronner, and R.M. Chambers. 2010. Conservation of diamondback terrapin and catch of blue crab potential benefits of reducing terrapin and fish bycatch and with a bycatch reduction device. Marine Ecology Progress Series reducing the capture of crabs in infrequently checked pots 409: 171–179. make a compelling case for enacting legislation to require Roosenburg, W. M. 1991. The diamondback terrapin: Habitat BRDs on crab pots in terrapin habitat. requirements, population dynamics, and opportunities for conservation. In New Perspectives in the Chesapeake System: A Research and Management and Partnership, 227–234. Acknowledgments Thanks to B. Bronner and T. Russell for field Solomons, MD: Chesapeake Research Consortium Pub. No. assistance and to the Chesapeake Program Cooperative Ecosystems 137. Studies Unit, J. Pulver and C. Wilson, DoD/Navy Chesapeake Bay Roosenburg, W.M. 2004. The impact of crab pot fisheries on Program, Navy Region Mid-Atlantic Environmental for logistical terrapin (Malaclemys terrapin) populations: Where are we and support. Funding for A. S. Morris, E.F. Dever, and S.M. Wilson came wheredoweneedtogo?InConservation and ecology of turtles from the W&M Monroe Scholars Program, a DTWG research grant, of the mid-Atlantic region: A symposium,ed.C.Swarth,W.M. and an HHMI Science Education Program grant to the College of Roosenburg, and E. Kiviat, 23–30. Salt Lake City: Biblomania. William and Mary, respectively. Roosenburg, W.M., and J.P. Green. 2000. Impact of a bycatch reduction device on diamondback terrapin and blue crab capture in crab pots. Ecological Applications 10: 882–889. References Roosenburg, W.M., W. Cresko, M. Modesitte, and M.B. Robbins. 1997. Diamondback terrapin (Malaclemys terrapin) mortality in crab pots. Conservation Biology 11: 1166–1172. Burger, J. 1977. Determinants of hatching success in the diamondback Schaffer, C., R.C. Wood, T.M. Norton, and R. Schaffer. 2008. terrapin, Malaclemys terrapin. American Midland Naturalist 97: Terrapins in the stew. Iguana 15: 78–85. 444–464. Seigel, R.A., and J.W. Gibbons. 1995. Workshop on the ecology, Butler, J.A., and G.L. Heinrich. 2007. The effectiveness of bycatch status, and management of the diamondback terrapin (Mala- reduction devices on crab pots at reducing capture and mortality clemys terrapin) Savannah River Ecology Laboratory, 2 August of diamondback terrapins (Malaclemys terrapin) in Florida. 1994: Final results and recommendations. Chelonian Conserva- Estuaries and Coasts 30: 179–185. tion and Biology 1: 241–243. Butler, J.A., G.L. Heinrich, and R.A. Seigel. 2006. Third workshop on Wood, R.C. 1997. The impact of commercial crab traps on northern the ecology, status and conservation of diamondback terrapins diamondback terrapins, Malaclemys terrapin terrapin.InPro- (Malaclemys terrapin): Results and recommendations. Chelonian ceedings: Conservation, Restoration, and Management of Tor- Conservation and Biology 5: 331–334. toises and Turtles–An International Conference, ed. J. Van Cagle, F.R. 1939. A system of marking turtles for future identification. Abbema, 21–27. New York: New York Turtle and Tortoise Copeia 1939: 170–173. Society.